Abstract
Background
Job demand-control (DC) and effort-reward-imbalance (ERI) are two commonly-used measures of work stress which are independently associated with health.
Aims
To test the hypothesis that DC and ERI have different and cumulative effects on health.
Methods
DC and ERI were assessed in the Hertfordshire Cohort Study. The characteristics and occupations of men and women reporting either or both work stresses were compared and the interaction of these with health status were explored.
Results
Complete data were available for 1,021 men and 753 women, reporting on their most recent or current job. 647 (63%) men and 444 (59%) women reported neither work stress whilst 103 (10%) men and 78 (10%) women reported both. Patterns of ERI and DC, alone and in combination, were different by type of occupation and by gender. Men reporting both work stresses (as compared with neither) were more likely to be single. Reported ERI with DC in the most recent or current job was associated with: poorer SF-36 physical function scores (OR 2.3 [95%CI 1.5-3.7] for men; OR 2.0 [95%CI 1.2-3.6] for women) and mental health scores (OR 2.8 [95%CI 1.8-4.4] for men; OR 3.1 [95%CI 1.8-5.3] for women). Moreover, average grip strength was 1.7kg (95%CI 0.2-3.3) lower among men who described both work stresses.
Conclusion
DC and ERI are two models of the psychosocial workplace environment which offer different but cumulative insight into the impacts of work on an individual’s psychological and physical health, particularly in a population sample.
Keywords: Occupation, demand-control, effort-reward imbalance, psychosocial workplace stressors
Introduction
Adverse psychosocial work conditions have been associated with impaired employee health cross-sectionally (1) and longitudinally (2,3). Occupational health researchers have tried to gain insight into the relationship between work stresses and employee health through conceptual frameworks. Two such frameworks which are commonly applied in research include Karasek’s job demand-control (DC) model (4) and Siegrist and colleagues’ effort-reward imbalance (ERI) model (5). The key principle behind the Karasek DC model is that control over the way an individual works might buffer the impact of job demands on occupational stress. Control over work is assessed by combining the answers to a series of questions about whether workers are able to choose or change the order of their tasks, their methods of work and their speed or rate of work, whether they have a say in the choice of their working partners and whether they can take a break when they wish. A composite indicator for the level of work demands (physical and psychological) is constructed by combining answers to questions that ask to what extent respondents have to work to tight deadlines and at high speed. Karasek and colleagues defined high DC as the adverse consequence which occurs in situations where high levels of perceived demand are not being matched by high levels of perceived control. It has been shown that DC is associated with a range of adverse impacts on: health, motivation and productivity (6,7,8).
Siegrist and colleagues focussed on the perceived reward, rather than the control structure, of work. Their theoretical model is based on the principle of social reciprocity, which implies mutual cooperative investments based on the norm of return expectancy where efforts are equalized by respective rewards. The model of effort-reward imbalance (ERI) claims that failed reciprocity in terms of high effort and low reward received is likely to elicit recurrent negative emotions and sustained stress responses in exposed people (9). Conversely, positive emotions evoked by appropriate social rewards promote well-being and health. Work effort is assessed by combining responses to questions about, for example, responsibility, interruptions, pressure to work overtime, changes in work situation. Perceived reward is calculated from responses to questions about the worker’s perceptions of: promotion prospects; job security; respect and prestige in the workplace. Although originally developed to study the relation between stress at work and cardiovascular disease, there is a growing body of literature showing that ERI is also associated with poorer subjective health status and common mental health conditions (9).
Although both models are associated with health outcomes, these two models have conceptually different origins and capture different domains of work stress. Therefore, it has recently been suggested that it would be worth studying the relative contribution of each model to wellbeing and health, in view of their differences and complementary aspects (10,11). The results of some, but not all (12), studies suggest that the combined effects of DC with ERI better predict poor health outcomes than each model separately (13,14,15,16,17). It is also recognised that the impact on a given health outcome may differ by occupation. For example, the results of one study suggested that DC appeared to have a larger impact on the risk of cardiovascular disease among blue-collar than white-collar workers (18). However, to date few studies have sampled ERI and DC amongst workers from different types of employment.
Therefore, this study explored the epidemiology and associations of ERI and DC within a cohort of adults at retirement age. We evaluated the overlap of ERI with DC among people doing different types of work and explored the characteristics of those with either or both work stress characteristics. Additionally, we explored the associations between high levels of DC or ERI, or both, with markers of health and wellbeing in this age group.
Methods
The Hertfordshire Cohort Study (HCS) has previously been described in detail elsewhere (19). In brief, the cohort comprises 1579 men and 1418 women born in Hertfordshire between 1931 and 1939 and who still lived there in 1998 to 2004. The study participants completed a nurse administered home interview during which the following characteristics were ascertained (see appendix 1 for details): marital status; own current or most recent full time occupation and husband’s current or most recent full time occupation for ever-married women; alcohol consumption; smoking status; age left full-time education; housing tenure; self-assessed health related quality of life using the short-form 36 (SF-36) questionnaire; medical history (including: stroke/transient ischaemic attack, or diabetes (out of pregnancy); typical angina (Rose chest pain questionnaire); history of coronary artery bypass graft or angioplasty; and details of all currently prescribed or over the counter medications, (coded to the British National Formulary). Participants subsequently attended a research clinic (see appendix 1 for details) at which height, weight, hand grip strength, and resting blood pressure were measured. A 2 hour fasting oral glucose tolerance test (OGTT) was performed using 75g anhydrous glucose and a standard 12-lead electrocardiogram was conducted. 1,021 men and 753 women completed a social health questionnaire which detailed employment related demand-control (20) and effort-reward imbalance (21) in the same manner as in the Whitehall II study (22–23), based upon their current or last held occupation.
Registrar General’s social class was coded from the 1990 OPCS Standard Occupational Classification (SOC90) unit group for occupation using computer assisted standard occupational coding (24).
The study had ethical approval from the East and North Hertfordshire NHS Research Ethics Committees and all participants gave written informed consent.
Participant characteristics were summarised using means and standard deviations or frequency and percentage distributions. Job demand and job control scores for current or most recent occupation were dichotomised on a sex-specific basis at their median values (which were: 58.3 and 50.0 years for men and women respectively for job demand score, and 71.3 and 57.4 years for job control score); these binary variables were cross-tabulated and study participants with high demand and low control scores were classified as reporting DC. The same method was used to identify study participants with ‘effort-reward imbalance’ (ERI) according to their job effort and reward scores (median values were: 18.8 and 12.5 for men and women respectively for job effort score, and 95.0 and 96.9 for job reward score). Finally, our principal four-level categorical exposure variable was coded by combining the DC and ERI variables as follows: ‘neither DC nor ERI’; ‘DC and ERI’; ‘DC only’ and ‘ERI only’. Study participants who reported ‘neither DC nor ERI’ were used as the reference group in all analyses.
Participant characteristics were described according to categories of DC/ERI and differences between groups were tested with Chi-squared tests or one-way analysis of variance as appropriate. Associations between DC/ERI and health outcomes were analysed using logistic regression (for binary outcomes). The normally distributed continuous variable, grip strength, was analysed as the estimated average difference in grip strength in the exposure category of interest in comparison with the nominated baseline group, calculated by multiple linear regression. Analyses were conducted with and without adjustment for the following potential confounders: age; age left full-time education; and SF-36 physical function or mental health scores. Subsequently, the same analyses were repeated with mutual adjustment to explore any amplification effects and whether there was a dominant effect from one stress. We also described DC/ERI according to commonly reported occupational groups. Analyses were conducted using Stata (release 14.2, Statacorp, College Station, Texas, USA) and men and women were analysed separately throughout.
Results
Table 1 describes participant characteristics according to categories of DC and ERI for the 1021 men and 753 women. 647 (63%) men and 444 (59%) women reported neither DC nor ERI in their current or most recent occupation; 103 (10%) men and 78 (10%) women reported both DC and ERI (Table 1). 658 (64%) men and 615 (82%) women reported that they had stopped work; the median (interquartile range [IQR]) number of years since stopping work was 4.6 (2.3,7.7) for men and 6.8 (3.8,11.5) for women.
Table 1. Characteristics of participants according to DC and effort-reward imbalance.
| Men (n=1021) | Women (n=753) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n(%) | n | Whole sample | Both ERI/DC | only DC | only ERI | neither | p | n | Whole sample | Both ERI/DC | only DC | only ERI | neither | p |
| 1021 | 103 | 65 | 206 | 647 | 753 | 78 | 83 | 148 | 444 | |||||
| Age (years, mean[SD]) | 1,021 | 65.4 (2.9) | 65.0 (2.9) | 65.1 (2.9) | 65.4 (2.7) | 65.5 (2.9) | NS | 753 | 66.1 (2.6) | 65.7 (2.6) | 66.0 (2.5) | 65.9 (2.6) | 66.2 (2.6) | NS |
| BMI (kg/m2) | ||||||||||||||
| < 18.5 (Underweight) | 1007 | 1 (0) | - | - | - | 1 (0) | NS | 746 | 6 (1) | - | 1 (1) | 2 (1) | 3 (1) | NS |
| 18.5 - 25 (Normal) | 295 (29) | 23 (23) | 22 (34) | 61 (30) | 189 (30) | 246 (33) | 26 (34) | 30 (36) | 42 (29.) | 148 (34) | ||||
| 25 - 30 (Overweight) | 504 (50) | 56 (56) | 31 (48) | 97 (48) | 320 (50) | 303 (41) | 30 (39) | 32 (39) | 67 (46) | 174 (40) | ||||
| > 30 (Obese) | 207 (21) | 22 (22) | 11 (17) | 46 (23) | 128 (20) | 191 (26) | 21 (27) | 20 (24) | 34 (24) | 116 (26) | ||||
| Marital status | ||||||||||||||
| single | 1,021 | 49 (5) | 9 (9) | 5 (8) | 12 (6) | 23 (4) | NS | 753 | 28 (4) | 6 (8) | 3 (4) | 3 (4) | 12 (3) | NS |
| ever married | 972 (95) | 94 (91) | 60 (92) | 194 (94) | 624 (97) | 725 (96) | 72 (92) | 80 (96) | 141 (95) | 432 (97) | ||||
| Social class | ||||||||||||||
| I-IIINM | 988 | 436 (44) | 39 (39) | 19 (30) | 109 (54) | 269 (43) | <0.01 | 753 | 325 (43) | 29 (37) | 28 (34) | 68 (46) | 200 (45) | NS |
| IIIM-V | 552 (56) | 62 (61) | 45 (70) | 92 (46) | 353 (57) | 428 (57) | 49 (63) | 55 (66) | 80 (54) | 244 (55) | ||||
| Alcohol consumption | ||||||||||||||
| Non-drinker / V Low | 1021 | 148 (15) | 16 (16) | 9 (14) | 31 (15) | 92 (14) | NS | 753 | 322 (43) | 36 (46) | 43 (52) | 57 (39) | 186 (42) | NS |
| Low | 413 (41) | 39 (38) | 32 (49) | 76 (37) | 266 (41) | 305 (41) | 29 (37) | 25 (30) | 59 (40) | 192 (43) | ||||
| Moderate | 241 (24) | 26 (25) | 11 (17) | 56 (27) | 148 (23) | 89 (12) | 10 (13) | 11 (13) | 18 (12) | 50 (11) | ||||
| F High / High | 219 (22) | 22 (21) | 13 (20) | 43 (21) | 141 (22) | 37 (5) | 3 (4) | 4 (5) | 14 (9) | 16 (4) | ||||
| Smoking status | ||||||||||||||
| Never | 1,021 | 353 (35) | 32 (31) | 21 (32) | 78 (38) | 222 (34) | NS | 752 | 449 (60) | 42 (54) | 47 (57) | 86 (58) | 274 (62) | NS |
| Ex | 531 (52) | 62 (60) | 39 (60) | 101 (49) | 329 (51) | 230 (31) | 31 (40) | 28 (34) | 47 (32) | 124 (28) | ||||
| Current | 137 (13) | 9 (9) | 5 (8) | 27 (13) | 96 (15) | 73 (10) | 5 (6) | 8 (10) | 15 (10) | 45 (10) | ||||
| Age left full time education | ||||||||||||||
| ≤14 years | 1,021 | 193 (19) | 18 (18) | 13 (20) | 26 (13) | 136 (21) | NS | 753 | 111 (15) | 11 (14) | 17 (21) | 14 (10) | 69 (16) | NS |
| ≥15 years | 828 (81) | 85 (83) | 52 (80) | 180 (87) | 511 (79) | 642 (85) | 67 (86) | 66 (80) | 134 (91) | 375 (85) | ||||
| House ownership | ||||||||||||||
| owned/mortgaged | 1,021 | 854 (84) | 81 (79) | 59 (91) | 171 (83) | 543 (84) | NS | 753 | 596 (79) | 61 (78) | 59 (71) | 122 (82) | 354 (80) | NS |
| rented/other | 167 (16) | 22 (21) | 6 (9) | 35 (17) | 104 (16) | 157 (21) | 17 (22) | 24 (29) | 26 (18) | 90 (20) | ||||
Summary statistics are frequencies and percentages unless stated otherwise; SD: standard deviation; NM: non-manual; M: manual; p: p-value; NS: non-significant, p>0.05
Univariate analyses identified some differences in participant characteristics according to DC and ERI (table 1). Men who reported DC and ERI were more likely to be single than their counterparts who reported neither. Men who reported ERI without DC were more likely to be of non-manual social class (I-IIINM), and were less likely to have left full-time education early (i.e. aged 14 years or younger), than men in other DC/ERI groups. There were no substantial differences in the participant characteristics included in table 1 according to DC/ERI among women.
Figure 1 shows the distribution of DC and ERI in the most commonly reported occupational groups for men and women separately. Among men, ERI without DC was reported most frequently by managers and administrators, production managers in manufacturing, and specialist managers; DC in combination with ERI was reported relatively most frequently among machine and plant operatives and those employed in woodworking trades. Among women, DC (with or without ERI) was relatively more common among those who worked in health and related occupations or who were machine and plant operatives; ERI without DC was reported most frequently by women who were teaching professionals.
Figure 1. Distribution of DC and effort-reward imbalance by occupational group (data shown for the most commonly reported occupations for men and women).
Table 2 details the prevalence of health conditions and measures of health status obtained from the SF-36 by gender for the study participants, as well as the mean grip strength and absolute scores for each of the component parts of the work stress measures (effort and reward scores, demand and decision latitude scores).
Table 2. Health outcomes and prevalence of health conditions, physical and mental health status and scores for effort/reward and demand and control by gender.
| n (%) | Men (n=1021) | Women (n=753) |
|---|---|---|
| SF36 physical functioning (PF) score (M=1020; W=753)* | 90 (80,95) | 85 (70,95) |
| SF-36 poor PF (lowest 5th of distribution) | 200 (20) | 143 (20) |
| SF36 mental health (MH) score (M=1020; W=753) * | 88 (76,92) | 80 (68,88) |
| SF-36 poor MH (lowest 5th of distribution) | 233 (23) | 153 (20) |
| Ever been told had a stroke (M=1009; W=750) | 44 (4) | 21 (3) |
| Definite IHD (M=993; W=726) | 137 (14) | 52 (7) |
| High BP/BP medication (M=1008; W=744) | 388 (39) | 284 (38) |
| Type II diabetes (M=940; W=707) | 113 (12) | 80 (11) |
| Grip strength (kg) (M=1008; W=744, Mean[SD]) | 44.3 (7.5) | 26.8 (5.5) |
| Job effort score (M=1021; W=753) * | 18.8 (6.3,31.3) | 12.5 (0,25) |
| Job reward score (M=1021; W=753) * | 95 (84.4,100) | 96.9 (90.6,100) |
| Job demand score (M=1021; W=753) * | 58.3 (41.7,66.7) | 50 (33.3,66.7) |
| Job decision latitude score (M=1021; W=753) * | 70.4 (56.5,82.4) | 57.4 (44.4,70.4) |
PF: physical function; MH: mental health; M: men; W: women; IHD: ischaemic heart disease; BP: blood pressure; kg: kilogramme
Descriptive statistics are frequencies and percentages unless indicated otherwise
Descriptive statistics are the median and inter-quartile range because the variable had a negatively skewed distribution
Table 3 describes the associations between DC and ERI and health status. We found no significant associations between DC/ERI and cardiovascular morbidity (stroke, ischaemic heart disease, hypertension) or type II diabetes among men or women. However, the odds of having poor SF-36 physical function scores were elevated among men and women who reported both DC and ERI in contrast with those who reported neither of these work stresses (OR 2.3 [95%CI 1.5,3.7] for men; OR 2.0 [95%CI 1.2-3.6] for women). The same pattern was evident for poor SF-36 mental health scores (OR 2.8 [95%CI 1.8-4.4] for men; OR 3.1 [95%CI 1.8-5.3] for women). In addition, average grip strength was 1.7kg (95%CI 0.2-3.3) lower among men who reported DC and ERI in contrast with those who reported neither DC or ERI. These results were unaltered by adjustment for age, age left full time education, marital status, and self-reported SF-36 mental or physical health score.
Table 3. Associations between DC (DC) and effort-reward imbalance (ERI) and health outcomes.
| Men | Women | |||||
|---|---|---|---|---|---|---|
| DC/ERI classification | n (%) | OR (95%CI) | n (%) | OR (95%CI) | ||
| Health Outcome | with health outcome | Crude | Fully adjusted ‡ | with health outcome | Crude | Fully adjusted ‡ |
| SF-36 poor PF (lowest 5th of distribution, M=1020; W=753) | ||||||
| Neither of them | 108 (17) | Reference | 68 (15) | Reference | ||
| Both of them | 33 (32) | 2.3 (1.5,3.7) | 2.1 (1.3,3.4) | 21 (27) | 2.0 (1.2,3.6) | 1.8 (1.0,3.2) |
| Only DC | 14 (22) | 1.4 (0.7,2.6) | 1.3 (0.7,2.6) | 19 (23) | 1.6 (0.9,2.9) | 1.5 (0.8,2.7) |
| Only ERI | 45 (22) | 1.4 (0.9,2.1) | 1.2 (0.8,1.8) | 35 (24) | 1.7 (1.1,2.7) | 1.6 (1.0,2.6) |
| SF-36 poor MH (lowest 5th of distribution, M=1020; W=753) | ||||||
| Neither of them | 111 (17.2) | Reference | 68 (15) | Reference | ||
| Both of them | 38 (37) | 2.8 (1.8,4.4) | 2.5 (1.5,3.9) | 28 (36) | 3.1 (1.8,5.3) | 2.9 (1.7,5.0) |
| Only DC | 15 (23) | 1.4 (0.8,2.7) | 1.4 (0.7,2.6) | 24 (29) | 2.2 (1.3,3.9) | 2.1 (1.2,3.7) |
| Only ERI | 69 (34) | 2.4 (1.7,3.5) | 2.4 (1.6,3.4) | 33 (22) | 1.6 (1.0,2.5) | 1.5 (0.9,2.4) |
| Ever been told had a stroke (M=1009; W=750) | ||||||
| Neither of them | 27 (4) | Reference | 14 (3) | Reference | ||
| Both of them | 8 (8) | 1.9 (0.8,4.4) | 1.9 (0.8,4.4) | 2 (3) | 0.8 (0.2,3.6) | 0.7 (0.1,3.2) |
| Only DC | 2 (3) | 0.7 (0.2,3.1) | 0.7 (0.2,3.1) | 2 (2) | 0.8 (0.2,3.4) | 0.7 (0.2,3.2) |
| Only ERI | 7 (3) | 0.8 (0.3,1.9) | 0.8 (0.3,1.8) | 3 (2) | 0.6 (0.2,2.3) | 0.6 (0.2,2.2) |
| Definite IHD (M=993; W=726) | ||||||
| Neither of them | 84 (13) | Reference | 30 (7) | Reference | ||
| Both of them | 15 (15) | 1.2 (0.7,2.1) | 1.0 (0.6,1.9) | 3 (4) | 0.6 (0.2,1.9) | 0.6 (0.2,2.0) |
| Only DC | 10 (16) | 1.3 (0.6,2.6) | 1.2 (0.6,2.5) | 8 (10) | 1.4 (0.6,3.2) | 1.4 (0.6,3.3) |
| Only ERI | 28 (14) | 1.1 (0.7,1.7) | 1.0 (0.6,1.6) | 11 (8) | 1.1 (0.5,2.3) | 1.1 (0.6,2.4) |
| High BP/BP medication (M=1008; W=744) | ||||||
| Neither of them | 246 (39) | Reference | 172 (39) | Reference | ||
| Both of them | 44 (43) | 1.2 (0.8,1.8) | 1.2 (0.8,1.8) | 26 (34) | 0.8 (0.5,1.3) | 0.8 (0.5,1.4) |
| Only DC | 20 (31) | 0.7 (0.4,1.3) | 0.7 (0.4,1.3) | 33 (40) | 1.0 (0.6,1.7) | 1.0 (0.6,1.7) |
| Only ERI | 78 (38) | 1.0 (0.7,1.4) | 0.9 (0.7,1.3) | 53 (37) | 0.9 (0.6,1.3) | 0.9 (0.6,1.4) |
| Type II diabetes (M=940; W=707) | ||||||
| Neither of them | 70 (12) | Reference | 49 (12) | Reference | ||
| Both of them | 11 (12) | 1.0 (0.5,2.0) | 1.0 (0.5,2.0) | 5 (7) | 0.6 (0.2,1.5) | 0.5 (0.2,1.4) |
| Only DC | 11 (19) | 1.8 (0.9,3.6) | 1.8 (0.9,3.7) | 8 (10) | 0.9 (0.4,1.9) | 0.9 (0.4,1.9) |
| Only ERI | 21 (11) | 1.0 (0.6,1.6) | 1.0 (0.6,1.6) | 18 (13) | 1.2 (0.6,2.1) | 1.2 (0.7,2.1) |
| Mean (SD) | Average difference in kg (95%CI) | Mean (SD) | Average difference in kg (95%CI) | |||
| Grip strength (kg) (M=1008; W=744) | ||||||
| Neither of them | 44.6 (7.6) | Reference | 27.2 (5.2) | Reference | ||
| Both of them | 42.8 (7.4) | -1.7 (-3.3,-0.2) | -1.8 (-3.4,-0.3) | 26.2 (5.6) | -1.1 (-2.4,0.3) | -0.9 (-2.2,0.5) |
| Only DC | 43.7 (6.7) | -0.9 (-2.8,1.1) | -1.0 (-2.9,0.8) | 25.6 (6.3) | -1.7 (-3.0,-0.4) | -1.6 (-2.9,-0.3) |
| Only ERI | 44.6 (7.6) | 0.03 (-1.1,1.2) | -0.01 (-1.2,1.2) | 26.7 (6.0) | -0.5 (-1.5,0.5) | -0.4 (-1.4,0.6) |
PF: physical function; MH: mental health; M: men; W: women; IHD: ischaemic heart disease; BP: blood pressure; kg: kilogramme;
OR: odds ratio; CI: confidence interval; ERI: effort-reward imbalance
Adjusted for age, age when left full time education, marital status, and MH, except for MH which was adjusted for PF
Men and women who reported ERI in the absence of DC had increased odds of poor SF-36 mental health scores in contrast with those who reported neither DC or ERI, although adjustment for age, age left full-time education, and physical function score attenuated the association among women. In addition, women who reported ERI without DC had increased odds of having poor physical function scores, and women who reported DC without ERI had increased odds of having poor mental health scores, in contrast with their counterparts who reported neither psychosocial occupational exposure (table 3).
These analyses were repeated with mutual adjustment (Table 4, Appendix 2). We found that the overall results were unaltered. Once again, exposure to both work stresses was associated with poor physical function, poorer mental health and lower grip strength and the cardiovascular morbidities and diabetes mellitus were not significantly associated.
Discussion
In this general population sample of retirement-age adults, we explored the frequency of occurrence of ERI and DC, and their associations with health measures among men and women in different occupations. We found that 10% of men and women reported both work stresses in their current or most recent occupation and that co-occurrence of both stresses was associated with poorer SF-36 physical function scores and mental health scores in comparison with individuals reporting neither job stress. Moreover, average grip strength was 1.7kg lower among men with DC and ERI as compared with men reporting neither job stress.
Our study had several limitations. Firstly, data on DC and ERI were only available for 1,021 men and 753 women rather than the full HCS baseline sample of 3,000 participants; this reduced sample size arose because the social health questionnaire was phased out during later stages of the fieldwork to allow time for detailed ascertainment of customary physical activity via a lengthy nurse-administered questionnaire. Secondly, 64% of the men and 82% of the women who participated in HCS reported that they no longer worked and were therefore recalling job stresses for their most recent job; recall bias could have affected our results if study participants who had experienced adverse health outcomes and poor wellbeing answered the questions about their work systematically differently. Whilst differential recall could arise by recent SF-36 mental health and physical function, such bias cannot explain associations with objective measurements of grip strength. Thirdly, overall levels of DC and ERI were low in this cohort. This is perhaps because participants were failing to recall adverse circumstances experienced in a previous job or perhaps that the respondents’ answers about their work after stopping reflect more global aspects of their current well-being, e.g. concerns over money or living conditions and less accurately assess past work stress. It is also possible that perceptions about work-related distress had faded over time since retirement. It is also conceivable that the ‘most recent’ job which many participants were thinking about when responding to the questionnaire was not particularly stressful perhaps because it was a “bridge” job (one that is part-time or less demanding undertaken at the end of a career). These relatively low levels of work stress may explain the lack of associations found in the current study with cardio-metabolic outcomes. Certainly, the results in Tables 3 and 4 show some (non-significant) trends towards increased associations in men between stroke and hypertension with both work stresses and diabetes and IHD with DC, consistent with the findings of other researchers. Despite this, it is perhaps more important that we found associations with grip strength and physical and mental function. Our results require replication in a study which more fully characterises accumulated job-strain and effort-reward imbalance longitudinally across an individual’s lifetime occupational history.
Other researchers have suggested a differential association of DC and ERI with health outcomes but we found few studies that have attempted to explore the pattern of DC and ERI within common occupational groups in a population sample. Patterns differed between genders and by occupation. For example, no female managers/proprietors reported ERI without DC but this was a frequent finding among male managers/administrators. Among female teaching professionals, we found high levels of ERI without DC. It is noteworthy that the DC model was developed in the 1960-70s industrial economy when many more jobs were in manufacturing whilst the ERI model might reflect stressors in a service sector economy. The current study includes people born in the 1930s who were providing their responses at retirement age but their careers may have included a variety of different jobs and the questionnaire only allowed us to assess their reported levels of job stress in their last or current job. Given the complexity of these concepts and the extent to which they are personality driven, it is perhaps not surprising that adults with different levels of educational attainment and different types of work perceive and report these stresses differently. Therefore, two workers doing seemingly identical work might report (and experience) very different levels of stress with lasting health consequences. This highlights a challenge for employers and public health physicians as these stressors will not be a ’fixed’ construct associated with specific occupations in the same way that physical work exposures may be characterised. From the point of view of prevention, job stress provides a significantly greater challenge.
Our results accord with the view that DC and ERI are two different models of the psychosocial workplace environment which offer different but complementary insight into the impacts of work on an individual’s psychological and physical health (10,11). Our finding that 10% of this age group report both DC and ERI and that the individuals reporting both factors have a doubled risk of poorer physical function and as much as trebled risk of poor mental health is important and could have major implications for sickness absence and health-related job loss. Grip strength is an excellent marker of muscle strength and it has been shown that weaker grip strength predicts adverse health outcomes including hospital admissions, falls, fractures and higher rates of mortality (27). The finding of a significant reduction in grip strength among men of retirement age reporting both ERI and DC implies an important impact of these stressors on muscle strength and function and therefore, on physical health and longevity at older ages.
In conclusion, DC and ERI provide two frameworks for understanding workplace psychosocial effects but which are complementary, and provide different information. In the HCS, older people who report both adverse profiles report poorer subjective physical and mental health status according to SF-36 scales but also show objective evidence of poorer health and physical function as measured by grip strength. We recommend that researchers consider use of both models in prospective studies in order to better understand potentially modifiable organisational interventions that might enhance health and wellbeing at work and after retirement.
Supplementary Material
Key points.
10% of men and women in this population study reported both ERI and DC in their current or most recent occupation
Men and women report ERI and DC differently even when doing jobs with similar titles
Jobs in which employees report both work stressors have greater negative impact on an individual’s subjective health and objective muscle strength than jobs in which one or the other, or neither, are reported.
Acknowledgments
The authors acknowledge the contribution of Professor Avan Aihie Sayer to this manuscript. Professor Sayer was a co-investigator in the Hertfordshire Cohort Study and was responsible for the inclusion of the effort-reward imbalance questions within the questionnaire. Prof Sayer also reviewed several drafts of the manuscript and made recommendations to improve the final draft.
Footnotes
Competing Interests: None
References
- 1.Lang J, Thomas JL, Bliese PD, Adler AB. Job demands and job performance: the mediating effect of psychological and physical strain and the moderating effect of role clarity. J Occup Health Psychol. 2007;12:116–24. doi: 10.1037/1076-8998.12.2.116. [DOI] [PubMed] [Google Scholar]
- 2.Theorell T, Hammarstrom A, Aronsson G, Bendz LT, et al. A systematic review including meta-analysis of work environment and depressive symptoms. BMC Public Health. 2015;15:738. doi: 10.1186/s12889-015-1954-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Theorell T, Jood K, Jarvholm LS, Vingard E, et al. A systematic review if studies in the contributions of the work environment to ischaemic heart disease development. Eur J Public Health. 2016;26:470–7. doi: 10.1093/eurpub/ckw025. Epub 2016 Mar 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Karasek RA, Theorell T. Healthy Work. New York: Basic Books; 1990. [Google Scholar]
- 5.Siegrist J, Siegrist K, Weber I. Sociological concepts in the etiology of chronic disease: the case of ischaemic heart disease. Soc Sci Med. 1986;22:247–53. doi: 10.1016/0277-9536(86)90073-0. [DOI] [PubMed] [Google Scholar]
- 6.Bongers PM, de Winter CR, Kompier MA, Hilderbrandt VH. Psychosoal factors at work and musculoskeletal disease. Scand J Work Environ Health. 1993;19:297–312. doi: 10.5271/sjweh.1470. [DOI] [PubMed] [Google Scholar]
- 7.Stansfeld S, Candy B. Psychosocial work environment and mental health – a meta-analytic review. Scand J Work Environ Health. 2006;32:443–62. doi: 10.5271/sjweh.1050. [DOI] [PubMed] [Google Scholar]
- 8.Belkic KL, Landsbergis PA, Schnall PL, et al. Is DC a major source of cardiovascular disease risk? A critical review of the empirical evidence with a clinical perspective. Scand J Work Environ Health. 2004;30:85–128. doi: 10.5271/sjweh.769. [DOI] [PubMed] [Google Scholar]
- 9.Van Vegchel N, de Jonge J, Bosma H, Schaufeli W. Reviewing the effort-reward imbalance model: drawing up the balance of 45 empirical studies. Soc Sci Med. 2005;60:1117–31. doi: 10.1016/j.socscimed.2004.06.043. [DOI] [PubMed] [Google Scholar]
- 10.Kasl SV. Measuring job stressors and studying the health impact of the work environment: an epidemiologic commentary. J Occup Health Psychol. 1998;3:390–401. doi: 10.1037//1076-8998.3.4.390. [DOI] [PubMed] [Google Scholar]
- 11.de Jonge J, Bosma H, Peter R, Siegrist J. DC, effort-reward imbalance and employee well-being: a large-scale cross-sectional study. Soc Sci Med. 2000;50:1317–27. doi: 10.1016/s0277-9536(99)00388-3. [DOI] [PubMed] [Google Scholar]
- 12.Lee SJ, Lee JH, Gillen M, Krause N. Job stress and work-related musculoskeletal symptoms among intensive care unit nurses: a comparison between job demand-control and effort-reward imbalance models. Am J Ind Med. 2014;57:214–21. doi: 10.1002/ajim.22274. Epub 2013 Oct 25. [DOI] [PubMed] [Google Scholar]
- 13.Calnan M, Wadsworth E, May M, Smith A, Wainwright D. DC, effort--reward imbalance, and stress at work: competing or complementary models? Scand J Public Health. 2004;32:84–93. doi: 10.1080/14034940310001668. [DOI] [PubMed] [Google Scholar]
- 14.Ostry AS, Kelly S, Demers PA, Mustard C, Hertzman C. A comparison between the effort-reward imbalance and demand control models. BMC Public Health. 2003;3:10. doi: 10.1186/1471-2458-3-10. Epub 2003 Feb 27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ota A, Masue T, Yasuda N, Tsutsumi A, et al. Psychosocial job characteristics and insomnia: a prospective cohort study using the Demand-Control-Support (DCS) and Effort-Reward Imbalance (ERI) job stress models. Sleep Med. 2009;10:1112–7. doi: 10.1016/j.sleep.2009.03.005. Epub 2009 May 21. [DOI] [PubMed] [Google Scholar]
- 16.Peter R, Siegrist J, Hallqvist J, Reuterwall C, Theorell T, SHEEP Study Group Psychosocial work environment and myocardial infarction: improving risk estimation by combining two complementary job stress models in the SHEEP Study. J Epidemiol Community Health. 2002;56:294–300. doi: 10.1136/jech.56.4.294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kivimäki M, Virtanen M, Elovainio M, Kouvonen A, et al. Work stress in the etiology of coronary heart disease--a meta-analysis. Scand J Work Environ Health. 2006;32:431–42. doi: 10.5271/sjweh.1049. [DOI] [PubMed] [Google Scholar]
- 18.Torén K, Schiöler L, Giang WK, Novak M, Söderberg M, Rosengren A. A longitudinal general population-based study of DC and risk for coronary heart disease and stroke in Swedish men. BMJ Open. 2014;4:e004355. doi: 10.1136/bmjopen-2013-004355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Syddall HE, Aihie Sayer A, Dennison EM, et al. Cohort profile: the Hertfordshire Cohort Study. Int J Epidemiol. 2005;34:1234–42. doi: 10.1093/ije/dyi127. [DOI] [PubMed] [Google Scholar]
- 20.Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q. 1979;24:285. doi: 10.1016/j.ijpsycho.2008.12.006. Epub 2008 Dec 14. [DOI] [Google Scholar]
- 21.Siegrist J. Adverse health effects of high effort - low reward conditions at work. Journal of Occupational Health Psychology. 1996;1:27–43. doi: 10.1037//1076-8998.1.1.27. [DOI] [PubMed] [Google Scholar]
- 22.Marmot M, Brunner E. Cohort Profile: the Whitehall II study. Int J Epidemiol. 2005;34:251–6. doi: 10.1093/ije/dyh372. [DOI] [PubMed] [Google Scholar]
- 23.Kuper H, Singh-Manoux A, Siegrist J, Marmot M. When reciprocity fails: effort-reward imbalance in relation to CHD and health functioning within the Whitehall II study. OEM. 2002;59:777–784. doi: 10.1136/oem.59.11.777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Office of Population Censuses and Surveys. Standard occupational classification, Vol 1 Structure and definition of major, minor and unit groups. London: HMSO; 1990. [Google Scholar]
- 25.Griep RH, Nobre AA, Alves MG, da Fonseca Mde J, Cardoso Lde O, et al. DC and unhealthy lifestyle: results from the baseline cohort study, Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) BMC Public Health. 2015;15:309. doi: 10.1186/s12889-015-1626-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Xu W, Yu H, Gao W, Guo L, Zeng L, Zhao Y. When job stress threatens Chinese workers: combination of job stress models can improve the risk estimation for coronary heart disease--the BADCAR study. J Occup Environ Med. 2011;53:771–5. doi: 10.1097/JOM.0b013e318222b0b5. [DOI] [PubMed] [Google Scholar]
- 27.Cooper R, Kuh D, Cooper C, et al. Objective measures of physical capability and subsequent health: a systematic review. Age Ageing. 2011;40:14–23. doi: 10.1093/ageing/afq117. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.